Advertisement putting management method and device

Abstract

The invention discloses an advertisement putting management method and device, and relates to the technical field of computers. A specific embodiment of the method comprises the steps of obtaining a first user set corresponding to a target advertisement commodity; screening out a second user set from the first user set according to a preset rule; respectively obtaining a first value estimation result and a second value estimation result of each user in the second user set according to the second user set, the first value estimation model and the second value estimation model; and according to the first value estimation result and / or the second value estimation result of each user, determining a target user set corresponding to the target advertisement commodity so as to perform advertisement putting on the target user set. According to the embodiment, intelligent delivery of logistics advertisements can be realized, value estimation is carried out on the advertisement flow from multiple angles, advertisement commodities and users are matched more accurately, the advertisement delivery efficiency and accuracy are improved, the advertisement conversion rate is improved, and the logistics advertisement delivery effect is improved.

Description

Technical Field

[0001] The present invention relates to the field of computer technology, and in particular, to a method and device for advertising placement management. Background Art

[0002] At present, in the logistics advertising industry, when placing advertisements, it mainly relies on a general delivery plan, and targets and mines populations according to marketing needs for large-scale placement. The general delivery plan for logistics advertisements mainly relies on manual experience. Moreover, for the needs of each advertiser, corresponding marketing strategies need to be formulated to select populations, resulting in low efficiency, low accuracy in selecting populations, and low conversion rates. Summary of the Invention

[0003] In view of this, embodiments of the present invention provide a method and device for advertising placement management, which can realize intelligent placement of logistics advertisements, estimate the value of advertising traffic from multiple perspectives, more accurately match advertising products and users, improve the efficiency and accuracy of advertising placement, and further improve the conversion rate of advertisements, thereby enhancing the effect of logistics advertising placement.

[0004] To achieve the above object, according to one aspect of the embodiments of the present invention, a method for advertising placement management is provided, including:

[0005] Obtaining a first user set corresponding to a target advertising product;

[0006] Screening out a second user set from the first user set according to a preset rule;

[0007] According to the second user set, a first value prediction model, and a second value prediction model, respectively obtaining a first value prediction result and a second value prediction result for each user in the second user set;

[0008] Determining a target user set corresponding to the target advertising product according to the first value prediction result and / or the second value prediction result of each user, so as to perform advertising placement on the target user set.

[0009] Optionally, obtaining a first user set corresponding to a target advertising product includes:

[0010] Obtaining a historical user set of the target advertising product;

[0011] Determining each similar advertising product corresponding to the target advertising product;

[0012] Obtaining a historical user set of each similar advertising product;

[0013] Determining the first user set according to the historical user set of the target advertising product and the historical user sets of each similar advertising product.

[0014] Optionally, each user in the historical user set corresponds to one or more behavior types; determining the first user set according to the historical user set of the target advertising product and the historical user set of each similar advertising product includes:

[0015] Determining a first user subset of each behavior type in the historical user set of the target advertising product;

[0016] Determining a second user subset of each behavior type in the historical user set of the similar advertising product;

[0017] Determining the first user set according to the first user subset and the second user subset of each behavior type.

[0018] Optionally, screening out a second user set from the first user set according to a preset rule includes:

[0019] Obtaining the user characteristics of each user in the first user set;

[0020] Inputting the user characteristics of each user into a population selection model to obtain the order placement probability of each user, and obtaining the second user set according to the order placement probability of each user;

[0021] Wherein, the population selection model is trained according to the user characteristics and order placement results of each user among multiple users.

[0022] Optionally, the first value prediction model is obtained according to the following method:

[0023] Obtaining multiple advertising product-user pairs and the store entry results of each advertising product-user pair; the advertising product-user pair is composed of the advertising product and the user who has searched for the store corresponding to the advertising product; the store entry results include the result of entering the store corresponding to the advertising product after searching and the result of not entering the store corresponding to the advertising product after searching;

[0024] Obtaining the sample characteristics of each advertising product-user pair;

[0025] Training the first value prediction model according to the sample characteristics and store entry results of each advertising product-user pair.

[0026] Optionally, the second value prediction model is obtained according to the following method:

[0027] Obtain multiple pairs of advertised products and users and the order placement results for each pair of advertised products and users; the pair of advertised products and users is composed of the advertised product and the user who enters the store corresponding to the advertised product, and the order placement result includes the result of placing an order after entering the store corresponding to the advertised product and the result of not placing an order after entering the store corresponding to the advertised product;

[0028] Obtain the sample features of each pair of advertised products and users;

[0029] Train the second value prediction model according to the sample features and order placement results of each pair of advertised products and users.

[0030] Optionally, the first value prediction result is the store entry conversion rate of each user in the second user set for the target advertised product; the second value prediction result is the order placement conversion rate of each user in the second user set for the target advertised product; determining the target user set corresponding to the target advertised product includes:

[0031] Sort each user in the second user set according to the store entry conversion rate or the second conversion rate or the product of the store entry conversion rate and the order placement conversion rate;

[0032] Determine the target user set from the second user set according to the sorting result.

[0033] According to still another aspect of the embodiments of the present invention, there is provided an apparatus for advertising placement management, including:

[0034] An acquisition module, which acquires a first user set corresponding to a target advertised product;

[0035] A screening module, which screens out a second user set from the first user set according to a preset rule;

[0036] A first determination module, which respectively obtains a first value prediction result and a second value prediction result of each user in the second user set according to the second user set, the first value prediction model, and the second value prediction model;

[0037] A second determination module, which determines a target user set corresponding to the target advertised product according to the first value prediction result and / or the second value prediction result of each user, so as to perform advertising placement on the target user set.

[0038] According to another aspect of the embodiments of the present invention, there is provided an electronic device, including:

[0039] One or more processors;

[0040] A storage device for storing one or more programs,

[0041] When the one or more programs are executed by the one or more processors, the one or more processors implement the method for advertising placement management provided by the present invention.

[0042] According to another aspect of an embodiment of the present invention, there is provided a computer-readable medium having a computer program stored thereon, and when the program is executed by a processor, the method for advertising placement management provided by the present invention is implemented.

[0043] One embodiment of the above invention has the following advantages or beneficial effects: In the method for advertising placement management, first, a first user set corresponding to a target advertising product is obtained, and then a second user set is screened out from the first user set according to a preset rule; the first value prediction model and the second value prediction model are used to predict the value of each user in the second user set, and according to the obtained first value prediction result and the second value prediction result, a target user set is determined from the second user set, and advertising is placed for the target user set. This method provides an intelligent placement solution for logistics advertisements, models the advertising traffic more refinedly from multiple perspectives, provides more value for advertising products and users, matches advertising products and users more precisely, can uniformly model advertising products and traffic, and has better generalization. This method accurately models and predicts the traffic value from multiple dimensions, determines high-value traffic for precise placement, improves the placement efficiency, precision rate and conversion rate, and enhances the user experience; through this method, a core evaluation index system for the logistics advertising industry can be established to guide the optimization of the logistics advertising placement effect.

[0044] The further effects of the above non-conventional optional manners will be described below in conjunction with specific embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] The drawings are used to better understand the present invention and do not constitute an improper limitation to the present invention. Among them:

[0046] Figure 1 is a schematic diagram of the main process of a method for advertising placement management according to an embodiment of the present invention;

[0047] Figure 2 is a schematic diagram of the main process of another method for advertising placement management according to an embodiment of the present invention;

[0048] Figure 3 is a schematic diagram of the main process of yet another method for advertising placement management according to an embodiment of the present invention;

[0049] Figure 4 is a schematic diagram of the main modules of a device for advertising placement management according to an embodiment of the present invention;

[0050] Figure 5 is an exemplary system architecture diagram to which embodiments of the present invention can be applied;

[0051] Figure 6 is a schematic structural diagram of a computer system of a terminal device or a server suitable for implementing embodiments of the present invention. Detailed implementation manners

[0052] The following describes exemplary embodiments of the present invention with reference to the accompanying drawings. Various details of the embodiments of the present invention are included to facilitate understanding, and they should be considered merely exemplary. Therefore, those of ordinary skill in the art should recognize that various changes and modifications can be made to the embodiments described herein without departing from the scope and spirit of the present invention. Similarly, descriptions of well-known functions and structures are omitted below for clarity and conciseness.

[0053] It should be noted that in the technical solutions of the present disclosure, in terms of the collection, acquisition, update, analysis, processing, use, transmission, storage, etc. of user personal information, they all comply with the provisions of relevant laws and regulations, are used for legal purposes, and do not violate public order and good customs. Necessary measures are taken for user personal information to prevent illegal access to user personal information data, and to maintain the security of user personal information, network security, and national security.

[0054] Figure 1 is a schematic diagram of the main process of a method for advertising placement management according to an embodiment of the present invention, as Figure 1 shown, the method for advertising placement management includes the following steps:

[0055] Step S101: Obtain a first user set corresponding to a target advertising product;

[0056] Step S102: Screen out a second user set from the first user set according to a preset rule;

[0057] Step S103: Obtain a first value prediction result and a second value prediction result of each user in the second user set according to the second user set, the first value prediction model, and the second value prediction model respectively;

[0058] Step S104: Determine a target user set corresponding to the target advertising product according to the first value prediction result and / or the second value prediction result of each user, so as to perform advertising placement on the target user set.

[0059] In the embodiments of the present invention, the method for advertising placement management is used to determine a target population for advertising placement in the logistics advertising industry, improve the accuracy rate of advertising placement, increase the conversion rate, and reduce the advertising placement cost.

[0060] In an embodiment of the present invention, first, a first user set corresponding to a target advertising product is obtained. The target advertising product is a merchant for which advertising is to be placed, and the merchant corresponds to a store in an e-commerce platform. Among them, the first user set corresponding to the target advertising product may include:

[0061] Obtain the historical user set of the target advertising product;

[0062] Determine each similar advertising product corresponding to the target advertising product;

[0063] Obtain the historical user set of each similar advertising product;

[0064] Determine the first user set according to the historical user set of the target advertising product and the historical user set of each similar advertising product.

[0065] In an embodiment of the present invention, a first-order collaborative filtering logic can be used to determine the first user set. Obtain the historical user set of the target advertising product, and use the historical user set as the characteristic feature of the target advertising product, which can be a set of historical users of one or more behavior types corresponding to the target advertising product. For example, the historical user set may include users who have searched, browsed, entered the store, structured, or purchased the store or product corresponding to the target advertising product. Since for similar advertising products, high-value users are also similar, by mining the user traffic of similar advertising products and recommending it to the target advertising product, traffic recall is achieved. Then, obtain the historical user set of each advertising product, and then determine the intersection of the historical user set of each advertising product and the historical user set of the target advertising product. The advertising product with the number of users in the intersection greater than or equal to the preset threshold is used as the similar advertising product corresponding to the target advertising product, that is, the more users in the intersection between two advertisements, the higher the similarity between the two advertising products; then, according to the historical user set of each similar advertising product and the historical user set of the target advertising product, determine the first user set. The first user set can be the set of users in the historical user set of the similar advertising product excluding the intersection.

[0066] In an embodiment of the present invention, each user in the historical user set corresponds to one or more behavior types; as Figure 2 shown, determining the first user set according to the historical user set of the target advertising product and the historical user set of each similar advertising product includes:

[0067] Step S201: Determine the first user subset of each behavior type in the historical user set of the target advertising product;

[0068] Step S202: Determine the second user subset of each behavior type in the historical user set of the similar advertising product;

[0069] Step S203: Determine the first user set according to the first user subset and the second user subset of each behavior type.

[0070] In the embodiment of the present invention, the historical user set of the target advertising product includes users corresponding to multiple behavior types. Determine the first user subset of each behavior type in the historical user set of the target advertising product, that is, determine the users who have search behavior, click behavior, store entry behavior, add-to-cart behavior, and purchase behavior for the store corresponding to the target advertising product respectively; determine the second user subset of each behavior type in the historical user set of each similar advertising product, that is, for each similar advertising product, determine the users who have search behavior, click behavior, store entry behavior, add-to-cart behavior, and purchase behavior for the store corresponding to the similar advertising product respectively; then for each behavior type, determine the intersection of the first user subset and the second user subset, and then calculate the difference set between the second user subset and the intersection. Determine the first user set according to the difference sets of various behavior types. For example, the combination of the difference sets of various behavior types can be used as the first user set, and duplicate users are removed during the combination process; or, after determining the intersection of the first user subset and the second user subset for each behavior type, according to the intersections corresponding to various behavior types, perform combination processing to obtain the first user set. For example, advertising product B is a similar advertising product corresponding to target advertising product A. Determine the user set C of users who have entered the store of target advertising product A among the users who have purchased the store of advertising product B, and use the user set E of users who have purchased advertising product B but have not entered the store of target advertising product A as potential valuable users of target advertising product A and recommend them to target advertising product A. Through this method, a large number of potential valuable users can be mined for the target advertising product. In the real scenario of logistics advertising, sampling this method can double the user traffic and improve the order placement conversion rate.

[0071] In the embodiment of the present invention, after determining the first user set corresponding to the target advertising product, filter out the second user set from the first user set according to a preset rule. Among them, the preset rule can be a randomly selected method, and each user can be randomly selected from the first user set according to a proportion to form the second user set; the preset rule can also be custom-set, such as users who have orders within the past year, that is, filter out the users who have orders within the past year from the first user set to form the second user set.

[0072] In the embodiment of the present invention, filtering out the second user set from the first user set according to a preset rule includes:

[0073] Obtain the user characteristics of each user in the first user set;

[0074] Input the user characteristics of each user into the population selection model to obtain the order placement probability of each user, and obtain the second user set according to the order placement probability of each user.

[0075] In an embodiment of the present invention, the preset rule may also be a population selection model, that is, the second user set is screened from the first user set according to the population selection model. First, determine the user characteristics of each user in the first user set, input the user characteristics into the population selection model, and the order placement probability of the user can be obtained, so as to obtain the order placement probability of each user in the first user set. Then, sort the users in the first user set in descending order according to the order placement probability, and determine the second user set according to the sorting result. For example, the second user set can be composed of the first N users before sorting, where N is a natural number greater than or equal to 1; or, the second user set is composed of users whose order placement probability is greater than or equal to a preset threshold.

[0076] In an embodiment of the present invention, the population selection mode is obtained according to the following method: obtain the user characteristics and order placement results of each user among multiple users, and then train a population selection model according to the user characteristics and order placement results of each user. Among them, multiple users who have searched, clicked, entered the store, or added the target advertising product A within the first preset time window can be obtained, and then the order placement results of these multiple users within the second preset time window. Then, train according to the user characteristics and order placement results of each user among the multiple users to obtain a population selection model. The population selection model can be used to predict whether each user in the first user set will place an order in a future period of time. User characteristics may include user portrait characteristics, recent activity, search and add-to-cart information, purchase frequency, periodic purchase and other characteristics. By constructing user characteristics from multiple dimensions, the accuracy of model prediction can be improved. The population selection model can be used to regularly and accurately estimate the order placement probability of users in a future period of time, so as to screen out users with a higher order placement probability, realize population selection, and thus recall active advertising traffic from high-value potential users and improve the activity rate. Among them, the structure of the population selection model can be a GBDT (Gradient Boosting Decision Tree) model.

[0077] In an embodiment of the present invention, after determining the second user set, that is, after recalling high-value active traffic for the target advertising product, further estimate the value of the traffic. The value of the advertising traffic for the target advertising product includes two dimensions, such as exposure to enter the store and enter the store to place an order. The first value estimation model and the second value estimation model can be used respectively to estimate the value of each user in the second user set, and the first value estimation result and the second value estimation result are obtained respectively. Use the first value estimation result and / or the second value estimation result to further screen the second user set to determine the target user set, and perform advertising placement for the target advertising product to improve the accuracy of advertising placement.

[0078] In the embodiment of the present invention, as Figure 3 shown, the first value prediction model is obtained according to the following method:

[0079] Step S301: Obtain a plurality of advertisement product-user pairs and the store entry results of each advertisement product-user pair; an advertisement product-user pair is composed of the advertisement product and a user who has searched for the store corresponding to the advertisement product; the store entry results include the result of entering the store corresponding to the advertisement product after the search and the result of not entering the store corresponding to the advertisement product after the search;

[0080] Step S302: Obtain the sample features of each advertisement product-user pair;

[0081] Step S303: Train a first value prediction model according to the sample features and store entry results of each advertisement product-user pair.

[0082] In the embodiment of the present invention, the first value prediction model can be used to predict the store entry value of each user in the second user set, that is, obtain the features of each target advertisement product-user pair in the second user set, input the features into the first value prediction model, and obtain the store entry value of each target advertisement product-user pair. When training the first value prediction model, within a preset time window, obtain a plurality of advertisement product-user pairs and the store entry results of each advertisement product-user pair. The advertisement product-user pair is regarded as a whole. For any advertisement product, multiple users who have searched for the store corresponding to the advertisement product can be selected, and each user can form an advertisement product-user pair with the advertisement product. When obtaining the training samples, the advertisement product-user pairs that the user enters the store after the search can be used as positive samples, and the advertisement product-user pairs that the user does not enter the store after the search can be used as negative samples. The full-scale samples are formed according to the positive samples and negative samples. The full-scale samples include each advertisement product-user pair and the store entry result of the advertisement product-user pair. The store entry results include entering the store after the search and not entering the store after the search; after generating the full-scale samples, sampling can be performed according to the ratio of positive samples to negative samples of 1:1 to obtain the training samples. Then obtain the sample features of each advertisement product-user pair. Since the value prediction model predicts the user traffic value of the advertisement product, the sample features can include advertisement product features, user features, and features of the advertisement product-user pair, which can include statistical features, attribute features, etc. For example, the advertisement product features can include the number of times the store corresponding to the advertisement product is searched, the user features can include the portrait features of the user, and the features of the advertisement product-user pair can include the number of times the user searches and purchases in the store corresponding to the advertisement product; train a first value prediction model according to the sample features and store entry results of each advertisement product-user pair. Among them, the structure of the first value prediction model can be a GBDT model using the sklean library (a general machine learning library).

[0083] In an embodiment of the present invention, after training the first value prediction model using training samples, the relationship between the actual conversion probability of the full sample and the conversion probability of the sampled training samples can be used to calibrate and post-process the first value prediction model to offset the influence of sampling for maintaining the balance of positive and negative samples during model modeling.

[0084] In an embodiment of the present invention, the second value prediction model is obtained according to the following method:

[0085] Obtain multiple advertisement product-user pairs and the order placement results of each advertisement product-user pair; an advertisement product-user pair is composed of the advertisement product and the user who enters the store corresponding to the advertisement product, and the order placement results include the result of placing an order after entering the store corresponding to the advertisement product and the result of not placing an order after entering the store corresponding to the advertisement product;

[0086] Obtain the sample features of each advertisement product-user pair;

[0087] Train the second value prediction model according to the sample features and order placement results of each advertisement product-user pair.

[0088] In an embodiment of the present invention, the second value prediction model can be used to estimate the order placement conversion value of each user in the second user set, that is, obtain the features of each target advertisement product-user pair in the second user set, input the features into the second value prediction model, and obtain the order placement conversion value of each target advertisement product-user pair. When training the second value prediction model, within a preset time window, obtain multiple advertisement product-user pairs and the order placement results of each advertisement product-user pair. For any advertisement product, multiple users who enter the store corresponding to the advertisement product can be selected, and each user can form an advertisement product-user pair with the advertisement product. When obtaining training samples, the advertisement product-user pairs with orders placed after the user enters the store can be used as positive samples, and the advertisement product-user pairs without orders placed after the user enters the store can be used as negative samples. The full sample is composed of positive samples and negative samples, and the full sample includes each advertisement product-user pair and the order placement result of the advertisement product-user pair. The order placement results include placing an order after entering the store and not placing an order after entering the store. After generating the full sample, positive and negative samples can be sampled in a ratio of 1:1 to obtain training samples. Then obtain the sample features of each advertisement product-user pair, and perform model training according to the sample features and order placement results of each advertisement product-user pair to obtain the second value prediction model. The second value prediction model can be a GBDT model using the sklean library.

[0089] In an embodiment of the present invention, after training the first value prediction model using training samples, the relationship between the actual conversion probability of the full sample and the conversion probability of the sampled training samples can be used to calibrate and post-process the first value prediction model to offset the influence of sampling for maintaining the balance of positive and negative samples during model modeling.

[0090] In an embodiment of the present invention, the first value prediction result is the in-store conversion rate of each user in the second user set for the target advertising product; the second value prediction result is the order placement conversion rate of each user in the second user set for the target advertising product; determining the target user set corresponding to the target advertising product includes:

[0091] Sort each user in the second user set according to the in-store conversion rate or the second conversion rate or the product of the in-store conversion rate and the order placement conversion rate;

[0092] Determine the target user set from the second user set according to the sorting result.

[0093] In an embodiment of the present invention, the first value prediction result is the in-store conversion rate, and the second value prediction result is the order placement conversion rate. According to the marketing requirements of the advertising product, each user in the second user set can be sorted according to the in-store conversion rate or the order placement conversion rate, or the product of the in-store conversion rate and the order placement conversion rate. The sorting can be performed in descending order, and then the top N users can be selected to form the target user set, where N is a natural number greater than or equal to 1, or users with an in-store conversion rate or an order placement conversion rate or the product of the two greater than or equal to a preset threshold can be selected to form the target user set. The logistics advertisement is delivered to the target user set, so that based on the traffic prediction, on the basis of the recalled active traffic for the target advertising product, the traffic with high predicted value is circled for delivery, realizing the intelligent delivery of the logistics advertisement and improving the conversion rate.

[0094] In an embodiment of the present invention, the logistics advertisement has values in two dimensions: in-store after search and order placement after in-store. Therefore, the in-store conversion rate and the order placement conversion rate can be used as the core indicators of the logistics advertisement, and combined with the ROI (Return On Investment) indicator of cost, a comprehensive and accurate logistics advertisement industry evaluation index system can be established.

[0095] The method for advertising placement management according to an embodiment of the present invention first obtains a first user set corresponding to a target advertising product, and then screens out a second user set from the first user set according to a preset rule; uses a first value prediction model and a second value prediction model to predict the value of each user in the second user set, and determines a target user set from the second user set according to the obtained first value prediction result and second value prediction result, and performs advertising placement on the target user set. This method provides an intelligent placement solution for logistics advertisements, models the advertising traffic more refinedly from multiple perspectives, provides greater value for advertising products and users, matches advertising products and users more accurately, can uniformly model advertising products and traffic, and has better generalization. This method first discovers potential users with high value, then uses a crowd selection model to improve the activity of recalled advertising traffic, accurately models and predicts the traffic value from multiple dimensions, determines high-value traffic for accurate placement, improves the placement efficiency, accuracy and conversion rate, and enhances the user experience; through this method, a core evaluation index system for the logistics advertising industry can be established to guide the optimization of the logistics advertising placement effect.

[0096] According to another aspect of the embodiment of the present invention, as Figure 4 shown, an advertising placement management device 400 is provided, including:

[0097] An acquisition module 401 for acquiring a first user set corresponding to a target advertising product;

[0098] A screening module 402 for screening out a second user set from the first user set according to a preset rule;

[0099] A first determination module 403 for respectively obtaining a first value prediction result and a second value prediction result of each user in the second user set according to the second user set, the first value prediction model and the second value prediction model;

[0100] A second determination module 404 for determining a target user set corresponding to the target advertising product according to the first value prediction result and / or the second value prediction result of each user, so as to perform advertising placement on the target user set.

[0101] In the embodiment of the present invention, the acquisition module 401 is further configured to: acquire the historical user set of the target advertising product; determine each similar advertising product corresponding to the target advertising product; acquire the historical user set of each similar advertising product; and determine the first user set according to the historical user set of the target advertising product and the historical user set of each similar advertising product.

[0102] In the embodiments of the present invention, each user in the historical user set corresponds to one or more behavior types; the obtaining module 401 is further configured to: determine a first user subset of each behavior type in the historical user set of the target advertising product; determine a second user subset of each behavior type in the historical user set of the similar advertising product; and determine a first user set according to the first user subset and the second user subset of each behavior type.

[0103] In the embodiments of the present invention, the screening module 402 is further configured to: obtain the user characteristics of each user in the first user set; input the user characteristics of each user into a population selection model to obtain the order placement probability of each user, and obtain a second user set according to the order placement probability of each user; wherein the population selection model is trained according to the user characteristics and order placement results of each user among multiple users.

[0104] In the embodiments of the present invention, the first determination module 403 is further configured to: obtain multiple advertising product-user pairs and the store entry results of each advertising product-user pair; an advertising product-user pair is composed of the advertising product and the user who has searched for the store corresponding to the advertising product; the store entry results include the result of entering the store corresponding to the advertising product after the search and the result of not entering the store corresponding to the advertising product after the search; obtain the sample characteristics of each advertising product-user pair; and train a first value prediction model according to the sample characteristics and store entry results of each advertising product-user pair.

[0105] In the embodiments of the present invention, the first determination module 403 is further configured to: obtain multiple advertising product-user pairs and the order placement results of each advertising product-user pair; an advertising product-user pair is composed of the advertising product and the user who has entered the store corresponding to the advertising product, and the order placement results include the result of placing an order after entering the store corresponding to the advertising product and the result of not placing an order after entering the store corresponding to the advertising product; obtain the sample characteristics of each advertising product-user pair; and train a second value prediction model according to the sample characteristics and order placement results of each advertising product-user pair.

[0106] In the embodiments of the present invention, the first value prediction result is the store entry conversion rate of each user in the second user set for the target advertising product; the second value prediction result is the order placement conversion rate of each user in the second user set for the target advertising product; the second determination module 404 is further configured to: sort each user in the second user set according to the store entry conversion rate or the second conversion rate or the product of the store entry conversion rate and the order placement conversion rate; and determine a target user set from the second user set according to the sorting result.

[0107] According to another aspect of an embodiment of the present invention, there is provided an electronic device, including: one or more processors; a storage device for storing one or more programs, which when executed by the one or more processors, cause the one or more processors to implement the method for advertising placement management provided by the present invention.

[0108] According to still another aspect of an embodiment of the present invention, there is provided a computer-readable medium having a computer program stored thereon, and when the program is executed by a processor, the method for advertising placement management provided by the present invention is implemented.

[0109] Figure 5 An exemplary system architecture 500 is shown to which the method for advertising placement management or the device for advertising placement management according to an embodiment of the present invention can be applied.

[0110] As Figure 5 shown, the system architecture 500 may include terminal devices 501, 502, 503, a network 504, and a server 505. The network 504 is used to provide a medium for a communication link between the terminal devices 501, 502, 503 and the server 505. The network 504 may include various connection types, such as wired, wireless communication links, or fiber optic cables, etc.

[0111] Users can use the terminal devices 501, 502, 503 to interact with the server 505 through the network 504 to receive or send messages, etc. Various communication client applications may be installed on the terminal devices 501, 502, 503, such as shopping applications, web browser applications, search applications, instant messaging tools, email clients, social platform software, etc. (only as examples).

[0112] The terminal devices 501, 502, 503 may be various electronic devices having a display screen and supporting web browsing, including but not limited to smart phones, tablet computers, laptop portable computers, and desktop computers, etc.

[0113] The server 505 may be a server providing various services, such as a background management server (only as an example) that supports shopping websites browsed by users using the terminal devices 501, 502, 503. The background management server may analyze and process data such as product information query requests received, and feedback the processing results (such as target push information, product information - only as examples) to the terminal devices.

[0114] It should be noted that the method for advertising placement management provided by the embodiments of the present invention is generally executed by the server 505. Correspondingly, the device for advertising placement management is generally disposed in the server 505.

[0115] It should be understood, Figure 5The numbers of the terminal devices, networks, and servers in [the above description] are merely illustrative. According to the implementation requirements, there can be any number of terminal devices, networks, and servers.

[0116] Reference is made below to Figure 6 , which shows a schematic structural diagram of a computer system 600 of a terminal device suitable for implementing the embodiments of the present invention. Figure 6 The terminal device shown is merely an example and should not impose any limitation on the functions and usage scope of the embodiments of the present invention.

[0117] As Figure 6 shown, the computer system 600 includes a central processing unit (CPU) 601, which can perform various appropriate actions and processes according to the program stored in the read-only memory (ROM) 602 or the program loaded from the storage section 608 into the random access memory (RAM) 603. In the RAM 603, various programs and data required for the operation of the system 600 are also stored. The CPU 601, ROM 602, and RAM 603 are connected to each other via a bus 604. An input / output (I / O) interface 605 is also connected to the bus 604.

[0118] The following components are connected to the I / O interface 605: an input section 606 including a keyboard, a mouse, etc.; an output section 607 including, for example, a cathode ray tube (CRT), a liquid crystal display (LCD), etc., and a speaker, etc.; a storage section 608 including a hard disk, etc.; and a communication section 609 including a network interface card such as a LAN card, a modem, etc. The communication section 609 performs communication processing via a network such as the Internet. A drive 610 is also connected to the I / O interface 605 as needed. A removable medium 611, such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc., is installed on the drive 610 as needed so that a computer program read from it can be installed into the storage section 608 as needed.

[0119] Specifically, according to the embodiments disclosed in the present invention, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, the embodiments disclosed in the present invention include a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program includes program codes for performing the methods shown in the flowcharts. In such an embodiment, the computer program can be downloaded and installed from the network via the communication section 609, and / or installed from the removable medium 611. When the computer program is executed by the central processing unit (CPU) 601, the above functions defined in the system of the present invention are executed.

[0120] It should be noted that the computer-readable medium shown in the present invention can be a computer-readable signal medium, a computer-readable storage medium, or any combination of the two. A computer-readable storage medium can be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination of the above. More specific examples of a computer-readable storage medium can include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present invention, a computer-readable storage medium can be any tangible medium that contains or stores a program that can be used by or in conjunction with an instruction execution system, apparatus, or device. In the present invention, a computer-readable signal medium can include a data signal propagated in a baseband or as part of a carrier wave, which carries computer-readable program code. Such a propagated data signal can take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. A computer-readable signal medium can also be any computer-readable medium other than a computer-readable storage medium, which can send, propagate, or transmit a program for use by or in conjunction with an instruction execution system, apparatus, or device. The program code contained on a computer-readable medium can be transmitted using any appropriate medium, including but not limited to: wireless, wire, optical fiber, RF, etc., or any suitable combination of the above.

[0121] The flowcharts and block diagrams in the accompanying drawings illustrate the possible architectures, functions, and operations of systems, methods, and computer program products according to various embodiments of the present invention. In this regard, each block in a flowchart or block diagram can represent a module, a program segment, or a part of code that contains one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions marked in the blocks can occur in a different order than that marked in the accompanying drawings. For example, two consecutive blocks shown can actually be executed substantially in parallel, and they can sometimes be executed in the reverse order, depending on the functions involved. It should also be noted that each block in a block diagram or flowchart, and combinations of blocks in a block diagram or flowchart, can be implemented by a dedicated hardware-based system for performing the specified functions or operations, or by a combination of dedicated hardware and computer instructions.

[0122] The modules involved in the embodiments of the present invention can be implemented in software or in hardware. The described modules can also be provided in a processor. For example, it can be described as: a processor includes an acquisition module, a screening module, a first determination module, and a second determination module. Among them, the names of these modules do not constitute a limitation to the module itself in some cases. For example, the acquisition module can also be described as "a module for acquiring a first user set corresponding to a target advertising product".

[0123] As another aspect, the present invention also provides a computer-readable medium, which can be included in the device described in the above embodiments; or it can exist alone without being assembled into the device. The above computer-readable medium carries one or more programs. When the above one or more programs are executed by the device, the device includes: acquiring a first user set corresponding to a target advertising product; screening out a second user set from the first user set according to a preset rule; respectively obtaining a first value prediction result and a second value prediction result of each user in the second user set according to the second user set, a first value prediction model, and a second value prediction model; and determining a target user set corresponding to the target advertising product according to the first value prediction result and / or the second value prediction result of each user, so as to perform advertisement placement on the target user set.

[0124] According to the technical solution of the embodiments of the present invention, the method for advertising placement management in the embodiments of the present invention first acquires a first user set corresponding to a target advertising product, and then screens out a second user set from the first user set according to a preset rule; uses a first value prediction model and a second value prediction model to perform value prediction on each user in the second user set, and determines a target user set from the second user set according to the obtained first value prediction result and second value prediction result, and performs advertisement placement on the target user set. This method provides an intelligent placement solution for logistics advertisements, models the advertisement traffic more refinedly from multiple perspectives, provides more value for advertising products and users, matches advertising products and users more accurately, can uniformly model advertising products and traffic, and has better generalization. This method first discovers potential users with high value, then uses a population selection model to improve the activity of recalled advertisement traffic, and accurately models and predicts the traffic value from multiple dimensions to determine high-value traffic for accurate placement, improving the placement efficiency, accuracy, and conversion rate, and enhancing the user experience; through this method, a core evaluation index system for the logistics advertising industry can be established to guide the optimization of the placement effect of logistics advertisements.

[0125] The above specific embodiments do not constitute a limitation on the protection scope of the present invention. Those skilled in the art should understand that various modifications, combinations, sub-combinations and substitutions can occur depending on design requirements and other factors. Any modifications, equivalent substitutions and improvements made within the spirit and principle of the present invention shall be included within the protection scope of the present invention.

Claims

1. A method for advertising placement management, characterized in that, it includes: Obtain a first user set corresponding to the target advertising product; Screen out a second user set from the first user set according to a preset rule; According to the second user set, the first value prediction model, and the second value prediction model, obtain the first value prediction result and the second value prediction result of each user in the second user set respectively; According to the first value prediction result and / or the second value prediction result of each user, determine a target user set corresponding to the target advertising product, so as to perform advertising placement on the target user set.

2. The method according to claim 1, characterized in that, Obtaining a first user set corresponding to the target advertising product includes: Obtain the historical user set of the target advertising product; Determine each similar advertising product corresponding to the target advertising product; Obtain the historical user set of each similar advertising product; Determine the first user set according to the historical user set of the target advertising product and the historical user set of each similar advertising product.

3. The method according to claim 2, characterized in that, Each user in the historical user set corresponds to one or more behavior types; Determining the first user set according to the historical user set of the target advertising product and the historical user set of each similar advertising product includes: Determine the first user subset of each behavior type in the historical user set of the target advertising product; Determine the second user subset of each behavior type in the historical user set of the similar advertising product; Determine the first user set according to the first user subset and the second user subset of each behavior type.

4. The method according to claim 1, characterized in that, Screening out a second user set from the first user set according to a preset rule includes: Obtain the user characteristics of each user in the first user set; Input the user characteristics of each user into the population selection model to obtain the order placement probability of each user, and obtain the second user set according to the order placement probability of each user; Wherein, the population selection model is trained according to the user characteristics and order placement results of each user among multiple users.

5. The method according to claim 1, characterized in that, The first value prediction model is obtained according to the following method: Obtain multiple advertising product - user pairs and the store entry results of each advertising product - user pair; the advertising product - user pair is composed of the advertising product and the user who has searched for the store corresponding to the advertising product; the store entry results include the result of entering the store corresponding to the advertising product after the search and the result of not entering the store corresponding to the advertising product after the search; Obtain the sample characteristics of each advertising product - user pair; Train the first value prediction model according to the sample characteristics and store entry results of each advertising product - user pair.

6. The method according to claim 1, characterized in that, The second value prediction model is obtained according to the following method: Obtain multiple pairs of advertised products and users and the order placement results for each pair of advertised products and users; the pair of advertised products and users is composed of the advertised product and the user who enters the store corresponding to the advertised product, and the order placement result includes the result of placing an order after entering the store corresponding to the advertised product and the result of not placing an order after entering the store corresponding to the advertised product; Obtain the sample features of each pair of advertised products and users; Train the second value prediction model according to the sample features and order placement results of each pair of advertised products and users.

7. According to the method described in claim 1, wherein, the first value prediction result is the store entry conversion rate of each user in the second user set for the target advertised product; the second value prediction result is the order placement conversion rate of each user in the second user set for the target advertised product; Determine the target user set corresponding to the target advertised product, including: Sort each user in the second user set according to the store entry conversion rate or the second conversion rate or the product of the store entry conversion rate and the order placement conversion rate; Determine the target user set from the second user set according to the sorting result.

8. An advertising placement management device, wherein, it includes: An acquisition module that acquires the first user set corresponding to the target advertised product; A screening module that screens out the second user set from the first user set according to a preset rule; A first determination module that respectively obtains the first value prediction result and the second value prediction result of each user in the second user set according to the second user set, the first value prediction model, and the second value prediction model; A second determination module that determines the target user set corresponding to the target advertised product according to the first value prediction result and / or the second value prediction result of each user, so as to perform advertising placement on the target user set.

9. An electronic device, wherein, it includes: One or more processors; A storage device for storing one or more programs, when the one or more programs are executed by the one or more processors, enabling the one or more processors to implement the method described in any one of claims 1-7.

10. A computer-readable medium, on which a computer program is stored, wherein, when the program is executed by a processor, it implements the method described in any one of claims 1-7.

Images